Magnetic Patterning by Electron Beam-Assisted Carbon Lithography

Subscriber access provided by UNIV OF DURHAM

Functional Nanostructured Materials (including low-D carbon)

Magnetic patterning by electron beam assisted carbon lithography Pietro Genoni, Francesca Genuzio, Tevfik Onur Mente#, Benito Santos, Alessandro Sala, Cristina Lenardi, and Andrea Locatelli ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.8b07485 • Publication Date (Web): 18 Jul 2018 Downloaded from http://pubs.acs.org on July 24, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Magnetic patterning by electron beam assisted carbon lithography Pietro Genoni,† Francesca Genuzio,‡ Tevfik Onur Mente¸s,‡ Benito Santos,‡ Alessandro Sala,‡,¶ Cristina Lenardi,† and Andrea Locatelli∗,‡ †CIMAINA, Dept. of Physics, Università degli Studi di Milano, via Celoria 16, I-20133 Milan, Italy ‡Elettra - Sincrotrone Trieste, S.S. 14 km 163.5 in AREA Science Park, Basovizza, I-34149 Trieste, Italy ¶Dept. of Physics, Università degli Studi di Trieste, via Valerio 2, I-34131 Trieste, Italy E-mail: [email protected]

1

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Abstract We report on the proof-of-principle of a scalable method for writing the magnetic state by electron-stimulated molecular dissociative adsorption on ultra-thin Co on Re(0001). Intense micro-focused low-energy electron beams are used to promote the formation of surface carbides and graphitic carbon through the fragmentation of carbon monoxide. Upon annealing at CO desorption temperature, carbon persists in the irradiated areas, whereas the clean surface is recovered elsewhere, giving origin to chemical patterns with nm-sharp edges. The accumulation of carbon is found to induce an inplane to out-of-plane spin reorientation transition in Co, manifested by the appearance of striped magnetic domains. Irradiation at doses in excess of 1000 L of CO followed by UHV annealing at 380 ◦ C determines the formation of a graphitic overlayer in the irradiated areas, under which Co exhibits out-of-plane magnetic anisotropy. Domains with opposite magnetization are here separated by chiral Neél walls. Our fabrication protocol adds lateral control to spin reorientation transitions, permitting to tune the magnetic anisotropy within arbitrary regions of mesoscopic size. We envisage applications in the nano-engineering of graphene-spaced stacks exhibiting desired magnetic state and properties.

Keywords: spin reorientation transition, cobalt thin films, magnetic lithography, patterning, chiral domain walls

2


Page 2 of 31

Page 3 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1

Introduction

The ability to create magnetic nanopatterns, i.e. controlled lateral modulations in the magnetic anisotropy of a continuous magnetic film, is key to the development of devices and sensors with novel architecture and functionality. 1 Magnetic patterning is often carried out by means of ion irradiation combined with conventional lithography and etching, and permits lateral resolution nowadays nearing the nanometer. 2 The effective magnetic anisotropy of the magnetic medium is controlled through the energy and fluence of the ions, which can induce structural transitions, 3,4 disorder at the film interfaces, 5–7 or change the chemical state of the material. 8 Irradiation and implantation methods, however, have a limited field of application, since they can produce irreversible structural damage. Varied techniques have been proposed as alternative, e.g. thermally assisted scanning probe lithography 9 or electron irradiation 10,11 to name just a few. The usage of e-beams for magnetic patterning was pioneered by Allenspach and coworkers. They studied ultra-thin Co films on Pt(111) and found out that irradiation with high energy electrons induces an in-plane to out-of-plane spin reorientation transition (SRT), which was attributed to a structural transformation of the magnetic medium. 10 Surprisingly, e-beam induced magnetic phenomena have remained substantially unexplored since then, with only a few exceptions noted. For instance, electron irradiation was found to induce the switching of a metastable state in Fe thin films grown on GaAs(110) producing an in-plane to in-plane SRT. This phenomenon was explained by invoking electronic excitations and ruling out adsorbate effects. 11 On the other hand, a recent study attributed an in-plane to in-plane SRT occurring in ultra-thin Co(110) to the electron stimulated irradiation and desorption of adsorbed carbon monoxide, 12 which permitted to demonstrate magnetic patterning at the mm scale. 13 Yet, a microscopic proof of this concept is still missing and its mechanisms certainly deserve further investigation. As a novel aspect pertinent to this subject, we address here the effect that stimulated molecular dissociative adsorption induces on thin-film magnetism, focusing on the simple 3



case of CO on cobalt. Indeed, electron-beam stimulated processes involving simple molecular adsorbates have attracted huge attention in the past decades, both experimentally and theoretically. 14,15 Yet, these studies mainly focused on the fundamental physical mechanisms of molecular fragmentation, leaving basically unexplored the territory of magnetism. In the following, we will describe a method to graft micron-sized carbon patterns inducing an inplane to out-of-plane SRT in few layer hcp Co on Re(0001). By irradiating the film surface in CO ambient using micro-focused beams, we will show that monolayer atomic carbon can be accumulated and, most remarkably, converted to a single graphitic layer upon a mild annealing treatment. As we will see, these carbon overlayers can greatly influence the magnetic anisotropy (MA) of the underlying Co thin-film. Our work is motivated by the current strong interest in magnetic heterostacks comprising a two-dimensional non-magnetic material as spacer layer, due to their relevance for spintronics. 1 As a prototypical 2D material, graphene is ideally suited for this type of application as it enhances perpendicular magnetic anisotropy in Co thin films, 16 indeed a very desirable characteristic for most magnetic memory devices. Notably, graphene-spaced magnetic stacks can be considered as the essential building blocks for the synthesis of ferromagnetic and antiferromagnetic materials with tunable magnetic properties. In this regard, antiferromagnetic exchange-coupling was recently demonstrated in Fe/graphene/Co layers supported on Ir(111). 17 As a further reason of interest, we note that graphene-Co is currently under the spotlight owing to the observation of chiral spin textures at graphene/ferromagnetic metal interface, their origin being attributed to a Rashba effect. 18 The capability to locally deposit 2D carbon layers is thus highly desirable. As will be shown, our method does not require high temperatures, which, together with lithographic capability, is a notable advantage over CVD techniques.

4


Page 4 of 31

Page 5 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


2

Experimental

All experiments were carried out at the Nanospectroscopy beamline of the Elettra Synchrotron (Trieste, Italy), using the spectroscopic photoemission and low energy electron microscope (SPELEEM). 19 This microscope allows us to correlate the magnetic state of the Co film, probed by x-ray magnetic circular dichroism photoemission electron microscopy (XMCD-PEEM), 20,21 to its surface stoichiometry and electronic structure, characterized by means of laterally resolved x-ray photoelectron spectroscopy (XPS) and microprobe angleresolved photoelectron spectroscopy (µ-ARPES), respectively. Crucial to our work, the SPELEEM can also be operated as a low energy electron microscope (LEEM), permitting us not only to monitor the growth and structural quality of the cobalt films, but also to perform micro-spot irradiation with its electron source. In LEEM, the specimen is probed with a micro-focused, collimated electron beam emitted by an LaB6 cathode. The microscope then collects the electrons that are elastically backscattered at the specimen surface, which can be imaged either in real or reciprocal space (see the Methods Section). The model system chosen for this work is affine to the thoroughly studied Co/Ru(0001), sharing with it the same substrate symmetry (hcp structure at room temperature, see the Methods Section) and similar lattice constant. It is important to note that ultra-thin Co films frequently display SRTs and other intriguing magnetic configurations. 20 In-plane to out-ofplane and out-of-plane to in-plane SRTs were observed for (hcp Co/Ru(0001), occurring at thicknesses of 2 and 3 atomic layers (AL), respectively. 22 Further, we note that the MA of ultra-thin Co films is known to be strongly influenced by the adsorption of CO and other gases, as has been repeatedly reported. 23–26 The choice of a heavy metal substrate, Re, was functional to obtaining strong and asymmetric exchange interactions at the interface, also known as Dzyaloshinskii-Moriya interaction (DMI). 27,28 Ultra-thin Co films of homogeneous thickness in the range 4 to 6 AL were grown as detailed in the Methods Section and Section 1 of the Supporting Information provided. At the thicknesses hereby considered, x-ray circular dichroic asymmetries up to 15% were ob5



served at the Co L3 edge. This behavior indicates predominantly in-plane MA for thicknesses greater or equal than 4 AL, considering that the dichroic effect is maximized in our set up for light impinging on the sample at an angle of incidence of 16◦ from the surface plane. Our observations are in agreement with previous measurements on closely related system Co/Ru(0001), which was found to exhibit in-plane MA at thicknesses equal or greater than 3 AL. 22 Yet, it has to be noted that XMCD-PEEM cannot easily quantify canted configurations, which were previously observed for Co films grown on W(110) at thickness in the range 3-8 monolayers (ML). 29 Nonetheless, based on image contrast considerations, we can safely exclude large tilt angles in the as grown Co film.

3

Concept of the patterning method

The protocol underlying chemo-magnetic patterning is illustrated in Figure 1. In the first step (top panel), the clean Co film (showing in-plane MA) is saturated with CO and then irradiated with an intense, micro-focused electron beam while being exposed to CO at room temperature. The irradiation with low energy electrons induces CO dissociative adsorption and desorption, as has been amply documented for varied transition metals. 15 As shown previously, the energy threshold for electron and photon CO dissociation is close to 15 eV; the largest cross section of the electron stimulated process is reached in the energy range 50 − 100 eV. 30,31 The e-beam irradiation shall be carried out in CO ambient at pressures in the range 10−9 − 10−6 mbar, so that molecular adsorption can counterbalance the effect of stimulated desorption. This permits the accumulation of C on the surface, as has been already reported by Lambert and Comrie. 32 We also note that, at the CO pressure considered here, molecular fragmentation mostly occurs on the surface, owing to the long resident time of CO following adsorption. Negligible dissociation occurs in the gas phase, the CO density being far too low to allow ionizing a significant amount of molecules. In the second step (bottom panel), UHV conditions are restored; the sample is then

6


Page 6 of 31

Page 7 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 1: Scheme of the grafting protocol used to produce chemo-magnetic patterns in ultrathin Co. (a, top) The film is exposed to CO at room temperature and irradiated with a focused electron beam, so that the dissociative adsorption of the molecule is induced. (b, bottom) The sample is annealed at 170 ◦ C to desorb molecularly adsorbed CO from the non-irradiated surface regions. annealed above CO desorption temperature, i.e. 170 ◦ C. 33 In this manner, molecularly adsorbed CO is removed from the non-irradiated regions, which will thus exhibit in-plane MA just like the as grown film. On the other hand, atomic carbon cannot be removed from the regions previously irradiated with the e-beam. In this respect, it is important to note that the bond strength of carbidic C to Co(0001) is much larger (6.47 eV/atom) than that of chemisorbed CO (1.46 eV/atom). 34 As will be shown below, both carbidic and graphitic carbon layers have a strong influence on the magnetic properties of the Co film, thereby inducing in-plane to out-of-plane SRTs. Clearly, the concept hereby illustrated can be generalized, for instance using other molecules in place of CO, provided they do not thermally dissociate during the annealing treatment.

7



4

Results and discussion

4.1

Irradiation at low CO doses

Figure 2 shows room temperature LEEM and XMCD-PEEM images of circular structures grafted on 4 AL thick Co films grown on Re(0001). All structures shown in (a) were fabricated by pre-dosing 2.5 L of CO before irradiation and then delivering further 4.5 L during irradiation, as previously detailed. The four smallest disks measure about 500 nm in diameter. LEEM imaging conditions were adjusted so that the irradiated spots appear brighter than their surroundings, exploiting differences in the low energy electron reflectivity that are related to the modified chemical composition of the topmost surface layer. The XMCDPEEM image in (b) illustrates the magnetic state of the region corresponding to the disks highlighted by the dashed contour in (a). Striped magnetic domains are clearly visible within the irradiated areas, indicating the onset of an in-plane to out-of-plane SRT. As seen in other experiments, the CO dose of 4.5 L is barely sufficient to induce the transition, so that the XMCD image depicts the magnetic state very close to the transition critical point. The LEEM and XMCD-PEEM images in Figure 2 (c-d) display a micron-sized disk that was printed by irradiating the cobalt surface for a longer time (a dose of about 10 L of CO was given during irradiation). LEEM demonstrates that disk exhibits quite sharp edges. A cross-sectional cut of the LEEM image across the disk edge is shown in Figure 2 (e), along with a step function fit that reveals an edge width of less than 30 nm. This broadening results from both the limited resolution of the microscope and the sample drift during irradiation. As can be seen in the XMCD-PEEM image in (d), the stripe pattern is fully established at this stage. An intensity profile perpendicular to the stripe direction reveals a period of about 120 nm, see (f). The emergence of a periodic stripe pattern is a clear, unambiguous sign of an in-plane to out-of-plane SRT. 35,36 The out-of-plane MA observed within the irradiated areas is confirmed by the decrease of the magnetic contrast due to the measurement geometry. 37 Overall, we

8


Page 8 of 31

Page 9 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 2: Images of chemo-magnetic patterns on a Co film with homogeneous thickness of 4 AL; all structures were grafted upon irradiation with 50 eV electrons at 2 · 10−8 mbar CO. The e-beam flux density on sample was 1.3 · 1016 e− cm−2 s−1 . (a) LEEM image at VS = 8 V of nine e-beam irradiated disks of diameter approx. 5, 1 and 0.5 µm. CO dose: 4.5 L;. (b) Co L3 edge XMCD-PEEM image of the region highlighted by the white box in (a) (c) LEEM image at VS = 9 V of an e-beam irradiated disk measuring approx. 1 µm in diameter; CO dose: 9.75 L; (d) XMCD-PEEM image of the same region at the Co L3 edge. (e) Intensity profile across the orange line in the LEEM image in (e) and fit using a step function convoluted with a Gaussian of FWHM of 30 nm. The dashed blue lines indicate the 15 − 85% distance between minimum and maximum intensity (f). Intensity profiles across the blue and orange dashed lines in the XMCD-PEEM image in (d). The plot highlights a difference in dichroic contrast of ∼ 3.5 between the irradiated disk and its surroundings. The magnetic stripes exhibit regular oscillations with a period of about 120 nm .

9



observed the SRT up to Co thicknesses of 6 AL. Our XMCD data shows clearly that the stripe period increases with increasing CO dose, as will be shown below, eventually leading to the formation of very large, stationary domains, as is expected when moving away from the critical point of the transition. 35 Temperature-dependent XMCD-PEEM measurements showed that the stripe patterns are visible up to about 140 ◦ C, a large enough value to ensure sufficient stability to the magnetic state observed at room temperature. The large XMCD contrast recorded outside the irradiated disks indicates that the magnetization is still oriented in-plane in the non-irradiated regions. The chemical state of the adlayer was investigated by probing the C 1s and O 1s core level emission inside and outside disks with diameter of 5 µm. The measurements were carried out at room temperature, after irradiation in CO ambient (PCO = 2.5 · 10−8 mbar) and UHV annealing. XPEEM images of an irradiated spot are shown in Figure 3 (a,b). They qualitatively demonstrate the different surface stoichiometry inside and outside the irradiated disk. These differences are quantitatively assessed by the spectra shown aside. As can be readily seen in Figure 3 (c), the intensity of the C 1s emission increases with the CO dose delivered during e-beam irradiation. In all spectra, the component at binding energy (BE) of 283.5 eV is clearly dominant and can be attributed to carbidic species. 38 Careful inspection of the spectrum obtained after 10 min. irradiation reveals a second component, centered at 284.9 eV, readily ascribed to the formation of sp2 carbon and thus graphitic in nature. 39 The adlayer coverage was estimated by evaluating the area of the measured C 1s spectrum and comparing it with that of single-layer graphene grown by high temperature CVD on a thick Co film supported on W(110). We determined a carbon coverage of 0.14 and 0.23 eMLgr (graphene equivalent monolayer) for the disks irradiated for 5 and 10 minutes, respectively. On the other hand, the non-irradiated part of the surface, still showing in-plane MA, exhibits a significantly lower C coverage, about 0.04 eMLgr , which is compatible with the surface being contaminated by the background pressure of CO and other hydrocarbons during the

10


Page 10 of 31

Page 11 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 3: (a,b) C 1s (hν=400 eV) and O 1s (hν=650 eV) XPEEM images of a 5 µm diameter disk, printed by e-beam irradiation in CO ambient. CO dose: 6.75 L. The images demonstrate carbon and oxygen accumulation within the e-beam irradiated region. (c,d) C 1s (hν=400 eV) and O 1s (hν=650 eV) spectra from the irradiated disks and the non-irradiated surface region surrounding it. The C enrichment reflects the CO dose. For explanation see text. time needed to prepare and perform the measurements. We thus estimate that the amount of C necessary to induce the SRT is very close to 0.1 eMLgr . Results for the different preparations reported in this study are summarized in Table 1. These findings demonstrate a clear tendency towards accumulation of carbon under x-ray irradiation in CO ambient. As a matter of fact, graphitization processes resulting from CO dissociation occur on all surfaces that are exposed to low energy electrons beams, extreme UV radiation and soft x-rays. 40 As well known by the x-ray optics community, they lead to detrimental effects affecting mirror reflectivity at and above the carbon K absorption edge. 41,42 Focusing now on O 1s spectra in Figure 3(d), the main peak (529.6 eV) is assigned to 11



Table 1: Summary of the results reported in this work. The C coverage θC , overlayer chemical state (type of carbon) and magnetic anisotropy (MA) of the irradiated and nonirradiated surface regions are listed as a function of the cobalt thickness (dCo ), CO dose delivered during e-beam irradiation (L) and final temperature and duration of the annealing treatment. Legend: → indicates in-plane MA, ↑ indicates out-of-plane MA. dCo (AL) 4 4-5 4-6 4-5 4.7

dose (L) 4.5 5.6 11.2 > 103 > 103

UHV annealing treatment θC 5’ 190 ◦ C 5’ 190 ◦ C 5’ 190 ◦ C ramp to 380 ◦ C 5 − 10’ 400 ◦ C

irradiated area (eMLgr ) C type 0.12 carbidic 0.14 carbidic 0.23 carbidic 0.77 graphitic 0.77 graphitic

non-irradiated surface MA θC (eMLgr ) C type MA ↑ 0.04 carbidic → ↑ 0.04 carbidic → ↑ 0.04 carbidic → ↑ 0.08 carbidic → ↑ Co islanding →

adsorbed atomic oxygen bound to cobalt forming CoO species. 43 We ascribe the tail at higher binding energy to the formation of a disordered oxygen adlayer, with multiple adsorption sites becoming populated. 44 The absence of emission at 532 eV, however, indicates that there is no molecular CO adsorbed on the surface, i.e. that CO has already undergone dissociative adsorption or desorption. Interestingly, the O 1s emission reaches an intensity maximum corresponding to the dose of 5 L of CO and notably decreases for higher CO doses. This observation is consistent with complementary photon stimulated desorption experiments that we performed, indicating that the desorption channel becomes progressively more important as C accumulates on the surface. The decrease of the O 1s signal is likely attributed to an attenuation effect due to the growing carbon layer. Finally, we estimate the ratio between the oxygen and carbon coverages by computing the areas of C 1s and O 1s peaks (in this case both spectra were measured at the same photon energy of 650 eV), keeping into account the photoionization cross-sections of the corresponding core-levels and the energy-dependent transmission of the spectrometer. We obtain a value of ∼ 1/3, which corresponds to an oxygen coverage of less than 0.1 eMLCo (monolayer equivalent of Co). We briefly mention that the magnetic patterns can be preserved from the action of external agents, e.g. oxidation, a crucial aspect for the fruition of our patterning method. This was achieved through the deposition of a thin Ag film (with thickness of ∼ 6 AL)

12


Page 12 of 31

Page 13 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


above Co. Ag was chosen because it maintains in-plane MA in the non-irradiated regions. 45 Layer-by-layer growth on cobalt is another notable advantage, resulting in the formation of high quality epitaxial layers. The capability of Ag-coated Co to withstand brief exposures (duration of few minutes) to ambient conditions while preserving the magnetic patterns is demonstrated in Section 2 of the Supporting Information provided.

4.2

Irradiation at high CO doses (> 103 L)

CO exposures of the order or in excess of 103 L result in a much larger amount of carbon being deposited on the surface. As noted above and also confirmed by XPS, carbon initially forms carbides and then accumulates in graphitic form. The disk-shaped structure Figure 4 (a) was fabricated by irradiating a 5 AL thick Co film with 60 eV electrons at a CO partial pressure of 1 · 10−6 mbar, delivering a total dose of 2.7 · 103 L. After restoring UHV conditions, the specimen was heated at rate of ∼ 1.35 ◦ C/s up to a maximum temperature of about 380 ◦ C. During treatment, we monitored the film structure with LEEM and observed a quick increase in the LEEM reflectivity at 8 eV at temperatures above 320 ◦ C. Not only this behavior reflects an improved ordering within the C overlayer, but also a chemical change, with carbon transforming from carbidic to graphitic type. The annealing treatment was stopped immediately after saturation of the LEEM intensity, in order to avoid dewetting and minimize islanding in cobalt. The transformations occurring in the adlayer and film upon thermal treatment are summarized in Scheme 1. The LEEM image in Figure 4 (a) proves the sharp morphology of the pattern. Importantly, we do not observe the opening of “holes” inside the disk, or islanding outside the irradiated areas, indicating that the cobalt has not undergone a lateral rearrangement. This is confirmed by measurements of the low energy electron reflectivity. As seen Figure 4 (b), the non-irradiated areas exhibit nearly identical characteristics before and after the treatment. Observation of quantum well resonances (see Section 1 in the Supporting information) indicates homogeneous characteristics across the surface and confirms that the initial thick13



Scheme 1: Diagram illustrating of the transformation of the adlayer and Co film upon thermal treatment. 1) the non-irradiated, CO covered surface is annealed to or above 200 ◦ C to desorb CO and recover the clean, as-grown surface. 2) the e-beam irradiated C-covered surface is ramped to 380 ◦ C; carbidic C transforms into graphitic C. Short treatments leave the morphology of the as-grown film unaffected, but prolonged annealing results in the formation of 3D Co islands in the non-irradiated regions. Islanding never occurs in the graphitized regions. ness of ∼ 5 AL is preserved. Prolonged annealing above 400 ◦ C results instead in some dewetting and islanding outside of the irradiated areas, but not inside. Consistent with previous data on Co under graphene on Ir(111), 46 the C-covered regions always appear laterally homogeneous, which can be attributed to graphene effectively hindering Co diffusion and islanding. As we could verify by XPS, further annealing does not improve the quality of the graphitic layer. XPEEM spectroscopy and imaging was used to locally assess the chemical state of a similarly fabricated micro-structure. As can be seen from the spectra plotted in Figure 4 (c), the C 1s emission strongly differs inside and outside the disk. The inset within the graph represents an XPEEM map of the C 1s emission and demonstrates well the accumulation of carbon inside the irradiated area. Within the disk, the C 1s emission is peaked at 284.9 eV, confirming the graphitic nature of the adlayer. 39 In fact, the line-shape and peak intensity (we estimate a coverage of 0.77 eMLgr ) compares well with that we recorded for CVD grown

14


Page 14 of 31

Page 15 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 4: (a) LEEM image of an e-beam irradiated 5 AL thick Co film after annealing to 380 ◦ C. A total CO dose of 2700 L was delivered during irradiation with 60 eV electrons. The small dark patches (top left) are surface regions which were not fully covered during the deposition of the first Co layer. Their thickness is 4 AL of Co. (b) LEEM-IV spectra inside and outside (blue and black curves respectively) the irradiated disk. Note that the low energy electron reflectivity of the surface does not change upon treatment (CO exposure followed by annealing for the non-irradiated regions).(c) XPEEM C 1s spectra from inside (blue line) and outside (black dashed line) another e-beam irradiated structure (a CO dose of 1350 L was delivered in this case); The C 1s spectrum (orange) of CVD graphene grown on the same substrate is shown for comparison. The XPEEM image in the inset represents the C map; Photon energy: 400 eV. (d) µ-ARPES data from inside the irradiated disk shown in (c). The graph is a cross sectional cut along the radial direction indicated by the dashed red line in pattern shown in the inset. The position of the K point is indicated by the black dashed line. The π band of rotationally incoherent graphene can be clearly recognized. Photon energy: 40 eV.

15



monolayer graphene supported on ultrathin Co/Re(0001) (also shown). Outside the disk, the C 1s emission is considerably weaker, indicating that a much smaller amount of carbon is adsorbed, about 0.08 eMLgr . The C 1s peak is here centered at 283.5 eV, fingerprint of carbide formation. This signal likely originates from residual CO adsorption and subsequent dissociation upon soft x-ray irradiation. Remarkably, no emission from the O 1s core level was detected inside or outside the irradiated area (not shown), suggesting that any residual adsorbed oxygen reacts with carbon during thermal treatment. The graphitic nature of the adlayer is demonstrated by the microprobe-ARPES data shown in Figure 4 (d) and confirmed by LEED. Both measurements indicate the formation of a rotationally incoherent graphene layer. Even if broadened, the π band of carbon is clearly visible in the momentum distribution curves, permitting us to estimate a Dirac energy close to 2.6 eV, in good agreement with literature data for single layer epitaxial graphene on Co(0001) 47 and the closely related system of graphene on Ni(111). 48 The magnetic state of the surface was investigated using XMCD-PEEM at the Co L3 edge. Figure 5 (a,b) show XMCD-PEEM images of the same structure shown in Figure 4 (a). Here, the contrast is purely magnetic and reveals a stripe pattern inside the irradiated area, confirming the enhancement of out-of-plane MA already observed at low carbon coverage. The bright and dark gray regions correspond to domains with the magnetization pointing up and down with respect to the film plane, respectively, as indicated by the symbols. The domain width has increased by a factor of about 3 compared to the case of low CO doses, suggesting that the magnetic state is slightly more distant from the SRT critical point. 49 Note also that the out-of-plane domains are separated by narrow regions were the contrast is enhanced (pointed by arrows in the Figure). This feature is well evidenced by intensity profile shown in Figure 5 (c). Such increase in dichroic contrast occurs solely when the domain walls are oriented at 90◦ from the x-ray beam. This means that the magnetization is aligned in-plane along the propagation direction of the photon beam, but perpendicular to the domain wall surface. As can be deduced from the arrows positioned in the top panel of

16


Page 16 of 31

Page 17 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 5 (c), the walls are of Neél type. Also, we observe that the magnetic contrast in the domain wall is dictated by the orientation of the out-of-plane domains, that is the domain walls are chiral. The observation of chirality reflects the occurrence of asymmetric DMI at the two interfaces of the film, a configuration that is known to occur in ternary superlattices 50,51 and MgO/Co/Pt stacks. We note that the domain walls observed in graphene/Co/Re exhibit opposite chirality with respect to those seen in MgO/Co/Pt, 52 which most likely points to the characteristics of the interaction with the different heavy metal. Nonetheless, graphene may have an crucial role in regulating the DMI, as has been recently proposed for graphene on Co/Ru. 18 To further investigate the magnetic properties of the graphene-coated Co, laterallyresolved XMCD spectra were acquired for out-of-plane domains with opposite magnetization. The computation of the magnetic circular dichroism sum rules (see Section 3 in the Supporting Information provided) permitted us to estimate the spin and orbital moments, µspin and µorb respectively, for which we find values of 1.7 and 0.16 µB /atom and a ratio µorb /µspin = 0.095. Our results are in excellent agreement with literature data for bulk hcp cobalt, 53 suggesting that the graphitic overlayer does not affect the atomic-scale magnetism of Co. Hence, we exclude the formation of a non-magnetic Co topmost layer, consistent with previous reports for CO-covered Co films. 23 In the present case, the perpendicular magnetic anisotropy (PMA) is most probably connected to electronic effects affecting the topmost Co layer after the chemisorption of atomic carbon. This interpretation is supported by previous work on the CO/Co/Pd(111) system, the occupation of specific adsorption sites being related to the occurrence of an in-plane to out-of-plane SRT in Co. 23,26 Graphene and varied carbon based molecules can affect the magnetic anisotropy of intercalated ferromagnetic layers in a similar manner. 54,55 In particular, graphene is known to stabilize out-of-plane magnetization in Co films on Ir(111) up to a thickness of 25 ˚ A. Related to this, ab-initio calculations demonstrate that this PMA enhancement is originated by the hybridization between Co dz2 and dyz orbitals, which is in turn induced by graphene p states. 16 Magneto-elastic effects are

17



Figure 5: (a) Co L3 edge XMCD-PEEM image of same surface region shown in Figure 4 (a). The irradiated disk exhibits a stripe pattern, indicating out-of-plane MA. (b) Magnified image of the region highlighted by the white dashed line in (a). As can be seen, the image contrast increases at the walls between domains with opposite magnetization. The red arrow represents the direction of the photon beam; the other symbols indicate the orientation of the magnetization vector. (c) Cross sectional intensity profiles taken along the blue and orange lines in (a). The blue curve illustrates the increase in dichroic contrast at the domain wall. The arrows mimic the orientation of the magnetization vector across the Neél type domain wall. Chirality is right handed. The orange contour exemplifies the larger image contrast that is observed in the regions exhibiting in-plane magnetic anisotropy. instead to be excluded, based on previous investigations on ultra-thin Co films covered by adsorbates 23,26 or capped by noble metals. 45 Clearly, our magnetic patterning method modifies the film magnetic state by forming chemical bonds at its vacuum interface. The film magnetic anisotropy is thus controlled through interfacial effects, whereas the bulk remains

18


Page 18 of 31

Page 19 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


structurally unaffected. This is a pronounced difference with ion irradiation patterning techniques, where the ions induce structural (e.g. disorder or a phase transition) or chemical transformations at much larger depths in the magnetic medium.

5

Conclusions

In conclusion, electron stimulated dissociation of CO was employed to graft chemo-magnetic micro-patterns with sharp edges (∼ 30 nm) in hcp cobalt films with thickness in the range 4 to 5 AL. These patterns exhibit a C rich surface and were found to induce PMA in the cobalt support, which has otherwise in-plane MA. The wavelength of out-of-plane striped magnetic domains was found to increase with the increasing amount of carbon deposited on the surface. Essentially, our method adds lateral control on the spin reorientation transition process, enabling to write patterns of arbitrary shape. Thus, one can aim at spatially confining the magnetization direction by simply controlling the amount of carbon that is deposited on the surface. Related to this, our results suggest that the interpretation of the early e-beam patterning experiments should be reconsidered keeping into account the effects of beam-induced dissociation of adspecies containing carbon. Most importantly, we demonstrated that prolonged e-beam irradiation in CO ambient results in the accumulation of ML amount of carbon on the surface, which can be converted to graphene upon subsequent annealing at 380 ◦ C. We envisage several applications of our method in fundamental studies on magnetism. For instance, one can fabricate artificial heterostacks comprising different ferromagnetic layers, where the magnetic coupling is tuned by an interposed graphene spacer. 17 Further work is clearly necessary to optimize conditions to grow atomically flat FM layers on graphene. Very promising results have already been produced in pulsed laser deposition experiments, which demonstrated close to perfect layer-by-layer growth for Co on graphene in the thickness range 3-5 AL. 56 As a first step in this direction, we focused here on graphene on ultra-thin Co on Re, which we consider

19



a prototypical building block for the fabrication of more complex stacks. We found that graphene enhances perpendicular MA in cobalt and, most interestingly, that the walls separating out-of-plane domains are chiral of Neél type. This suggests that skyrmions can be nucleated and stabilized in the Co film under appropriate conditions. In this respect, the lithographic capabilities of our method are particularly appealing, as they permit to exploit confinement effects without destroying the lateral continuity of the Co film. As a final point, we note that EUV radiation or soft-x-rays can be used in place of electrons, since CO molecular dissociation is also stimulated by photons. The use of masks or advanced e-beam lithography methods 57 can further improve the lateral resolution as well as the printing speed, adding versatility and true scalability to the printing protocol.

6

Methods section

SPELEEM: all experiments hereby described were carried out with the spectroscopic photoemission and low energy electron microscope (SPELEEM, Elmitec GmbH) at the Nanospectroscopy beamline of the Elettra storage ring in Trieste (Italy). In LEEM, the specimen is illuminated with a micro-focused, collimated electron beam emitted by an LaB6 cathode. The beam diameter on sample is about 80 µm. The beam energy, varied in the range 0 − 750 eV, is regulated by applying a bias voltage (start voltage VS ) to the sample. Imaging is possible in either real or reciprocal space, with three illumination apertures permitting to restrict the e-beam footprint to a diameter of 5 µm, 1 µm or 500 nm. When the beamline x-rays are used, the SPELEEM can implement laterally resolved versions of xray absorption spectroscopy and related methods, namely x-ray magnetic circular dichroism (XMCD-PEEM). Thanks to a bandpass energy filter, the SPELEEM also allows spectral imaging measurements (XPEEM), reaching lateral resolution below 30 nm and energy resolution of about 300 meV. Microprobe photoemission spectroscopy (e.g. µ-XPS) operation is also possible, which enables fast acquisition and energy resolution as good as ∼ 110 meV. 19

20


Page 20 of 31

Page 21 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


These measurements are restricted to an area of about 2 microns in diameter. Depending on the flux available, the binding energies were referenced to the Fermi level or the Re 4f core level emission from the bulk metal. XMCD-PEEM: x-ray magnetic circular dichroism (XMCD) is based on the dependence of the x-ray absorption intensity on the relative orientation of the magnetization and the x-ray polarization. In x-ray microscopy, this effect is used to image magnetic domains in thin ferromagnetic layers and nanostructures. 21 The XMCD image intensity was calculated −I+ , where I− and I+ represent the pixel intensities using the following algebra: IXM CD = II−− +I +

of the PEEM images acquired with negative and positive helicities of the photon beam. The image contrast is maximum when the photon helicity and the magnetization directions are parallel and anti-parallel. Due to grazing incidence of the photon beam at an angle of 16◦ with respect to the surface plane, the dichroic contrast recorded for in-plane magnetization is up to about 3.5 times larger than for the case of out-of-plane magnetization. Sample preparation: The Re(0001) single crystal (Surface Preparation Labs) was cleaned by Ar+ sputtering at 2 kV followed by repeated temperature cycling between 400 and 800 ◦ C at oxygen pressure of 2 · 10−6 mbar. After the treatment, residual oxygen was desorbed with a flash to 1700 ◦ C in UHV. Before experiments, no impurities could be detected in XPS. µ-LEED on a single terrace showed a sharp (1×1) pattern with three-fold symmetry. No evidence of oxygen-related half-order spots was seen, even diffuse ones. LEEM showed uniform contrast across terraces and no evidence of islanding, thus excluding the presence of contaminants. Co growth: All experiments were carried out at a base pressure of less than 1.5 · 10−10 mbar. Co was dosed using a commercial e-beam evaporator (Omicron Focus EMF3), sublimated from high-purity rods (Goodfellow ltd) at a maximum deposition rate of about 0.1 ML/min. The specimen temperature was measured using a C type thermocouple (W96 Re4 W74 Re26 ), spot-welded on the Mo ring supporting the sample. The first Co layer was deposited at 350 ◦ C, which warrants step-flow growth conditions. In this manner, defect-free

21



Page 22 of 31

pseudomorphic Co monolayers (ML) could be obtained. To achieve a uniform thickness of the Co film as well as prevent alloying with the Re substrate, the following layers were grown at room temperature, which results in the development of the 10 × 10 moiré structure, indicating relaxation of the lattice constant to the bulk value. 58 Post-growth annealing at 250 ◦ C was carried out to reduce surface roughness. LEEM measurements in the dark-field mode demonstrated that, after thermal treatment, Co retains the hcp structure. 59,60

7

Supporting Information

The Supporting Information is available free-of-charge on the ACS Publication website. The SI describes the methods for film thickness determination, capping layers for magnetic pattern preservation, the XMCD magnetometry results for graphene/5ALCo/Re.

Acknowledgement A.L. and T.O.M. thank CERIC-ERIC for support through the internal project MAG-ALCHEMI. We thank M. Kiskinova and E. Bauer for illuminating discussions and critical reading of this manuscript.

22


Page 23 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


References (1) Sander, D. et al. The 2017 Magnetism Roadmap. J. Phys. D: Appl. Phys. 2017, 50, 363001. (2) Randolph, S. J.; Fowlkes, J. D.; Rack, P. D. Focused, Nanoscale Electron-beam-induced Deposition and Etching. Crit. Rev. Solid State 2006, 31, 55–89. (3) Rupp, W.; Biedermann, A.; Kamenik, B.; Ritter, R.; Klein, C.; Platzgummer, E.; Schmid, M.; Varga, P. Ion-beam Induced fcc-bcc Transition in Ultrathin Fe Films for Ferromagnetic Patterning. Appl. Phys. Lett. 2008, 93, 063102. (4) Zaman, S. S.; Dvoˇràk, P.; Ritter, R.; Buchsbaum, A.; Stickler, D.; Oepen, H. P.; Schmid, M.; Varga, P. In-situ Magnetic Nano-patterning of Fe Films Grown on Cu(100). J. Appl. Phys. 2011, 110, 024309. (5) Chappert, C.; Bernas, H.; Ferré, J.; Kottler, V.; Jamet, J.-P.; Chen, Y.; Cambril, E.; Devolder, T.; Rousseaux, F.; Mathet, V.; Launois, H. Planar Patterned Magnetic Media Obtained by Ion Irradiation. Sci. 1998, 280, 1919–1922. ¨ (6) Bali, R.; Wintz, S.; Meutzner, F.; H¨ ubner, R.; Boucher, R.; Unal, A. A.; Valencia, S.; Neudert, A.; Potzger, K.; Bauch, J.; Kronast, F.; Facsko, S.; Lindner, J.; Fassbender, J. Printing Nearly-discrete Magnetic Patterns Using Chemical Disorder Induced Ferromagnetism. Nano Lett. 2014, 14, 435–441. (7) Hlawacek, G.; Bali, R.; Röder, F.; Aleksandrov, Y.; Semisalova, A.; Wintz, S.; Wagner, K.; Schultheiss, H.; Facsko, S.; Fassbender, J.; et al., Tailoring Magnetic Nanostructures with Neon in the Ion Microscope. Microscopy and Microanalysis 2016, 22, 17161717. (8) Kim, S.; Lee, S.; Ko, J.; Son, J.; Kim, M.; Kang, S.; Hong, J. Nanoscale Patterning

23



of Complex Magnetic Nanostructures by Reduction with Low-energy Protons. Nature Nanotech. 2012, 7, 567–571. (9) Albisetti, E.; Petti, D.; Pancaldi, M.; Madami, M.; Tacchi, S.; Curtis, J.; King, W.; Papp, A.; Csaba, G.; Porod, W.; Vavassori, P.; Riedo, E.; Bertacco, R. Nanopatterning Reconfigurable Magnetic Landscapes via Thermally Assisted Scanning Probe Lithography. Nature Nanotech. 2016, 11, 545–551, cited By 25. (10) Allenspach, R.; Bischof, A.; D¨ urig, U.; Gr¨ utter, P. Local Modification of Magnetic Properties by an Electron Beam. Appl. Phys. Lett. 1998, 73, 3598–3600. (11) Monchesky, T. L.; Unguris, J.; Celotta, R. J. Electron Beam Stimulated Spin Reorientation. J. Appl. Phys. 2003, 93, 8241–8243. (12) Reeve, R. M.; Chin, S.-L.; Kopper, K.-P.; Ionescu, A.; Barnes, C. H. W. Experimental and Theoretical Study of Electron-beam-induced Spin-reorientation Transition Reversal in the CO/Co(110) System. J. Phys. D: Appl. Phys. 2012, 45, 275003. (13) Reeve, R.; Chin, S. L.; Kopper, K. P.; Ionescu, A.; Barnes, C. H. W. Electron Beam Tuning of the Magnetic Anisotropy in Co/Cu(110) Films. IEEE Trans. Magn. 2011, 47, 1554–1558. (14) Avouris, P.; Walkup, R. E. Fundamental Mechanisms of Desorption and Fragmentation Induced by Electronic Transitions at Surfaces. Ann. Rev. Phys. Chem. 1989, 40, 173– 206. (15) Ramsier, R.; Jr., J. Y. Electron Stimulated Desorption: Principles and Applications. Surf. Sci. Rep. 1991, 12, 246–378. (16) Yang, H.; Vu, A. D.; Hallal, A.; Rougemaille, N.; Coraux, J.; Chen, G.; Schmid, A. K.; Chshiev, M. Anatomy and Giant Enhancement of the Perpendicular Magnetic Anisotropy of Cobalt-Graphene Heterostructures. Nano Lett. 2016, 16, 145–151. 24


Page 24 of 31

Page 25 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(17) Gargiani, P.; Cuadrado, R.; Vasili, H. B.; Pruneda, M.; Valvidares, M. Graphene-based Synthetic Antiferromagnets and Ferrimagnets. Nature Comm. 2017, 699, 1 – 9. (18) Yang, H.; Chen, G.; Cotta, A. A. C.; N’Diaye, A. T.; Nikolaev, S. A.; Soares, E. A.; Macedo, W. A. A.; Schmid, A. K.; Fert, A.; Chshiev, M. Significant DzyaloshinskiiMoriya Interaction at Graphene-Ferromagnet Interfaces due to Rashba Effect. ArXiv e-prints 2017, (19) Mente¸s, T.; Zamborlini, G.; Sala, A.; Locatelli, A. Cathode Lens Spectromicroscopy: Methodology and Applications. Beilstein J. Nanotech. 2014, 5, 1873–1886. (20) Bauer, E. Surface Microscopy with Low Energy Electrons; Springer: New York, 2014; pp 409–464. (21) Cheng, X. M.; Keavney, D. J. Studies of Nanomagnetism Using Synchrotron-based Xray Photoemission Electron Microscopy (X-PEEM). Rep. Prog. Phys. 2012, 75, 026501. (22) El Gabaly, F.; Gallego, S.; Mu˜ noz, C.; Szunyogh, L.; Weinberger, P.; Klein, C.; Schmid, A. K.; McCarty, K. F.; de la Figuera, J. Imaging Spin-Reorientation Transitions in Consecutive Atomic Co Layers on Ru(0001). Phys. Rev. Lett. 2006, 96, 147202. (23) Matsumura, D.; Yokoyama, T.; Amemiya, K.; Kitagawa, S.; Ohta, T. X-ray Magnetic Circular Dichroism Study of Spin Reorientation Transitions of Magnetic Thin Films Induced by Surface Chemisorption. Phys. Rev. B 2002, 66, 024402. (24) Yokoyama, T.; Matsumura, D.; Amemiya, K.; Kitagawa, S.; Suzuki, N.; Ohta, T. Spin Reorientation Transitions of Ultrathin Co/Pd(111) Films Induced by Chemisorption: X-ray Magnetic Circular Dichroism Study. J. Phys. Condens. Matter 2003, 15, S537. (25) Matsumura, D.; Amemiya, K.; Kitagawa, S.; Shimada, T.; Abe, H.; Ohta, T.; Watanabe, H.; Yokoyama, T. Surface Structure of CO on CoPd(111) Magnetic Thin Films

25



and its Effect on the Spin Reorientation Transition of the Film. Phys. Rev. B 2006, 73, 174423. (26) Matsumura, D.; Nakagawa, T.; Watanabe, H.; Abe, H.; Amemiya, K.; Ohta, T.; Yokoyama, T. Magnetization process of Co/Pd(111) thin films: Chemisorption-induced spin-reorientation transition. Surface Science 2008, 602, 1999 – 2003. (27) Dzyaloshinskii, I. E. A Thermodynamic Theory of Weak Ferromagnetism of Antiferromagnetics. J. Phys. Chem. Solids 1958, 4, 241255. (28) Moriya, T. Anisotropic Superexchange Interaction and Weak Ferromagnetism. Phys. Rev. 1960, 120, 9198. (29) Duden, T.; Bauer, E. Magnetization Wrinkle in Thin Ferromagnetic Films. Phys. Rev. Lett. 1996, 77, 2308–2311. (30) Itikawa, Y. Cross Sections for Electron Collisions with Carbon Monoxide. J. Phys. Chem. Ref. Data 2015, 44, 013105. (31) Ramaker, D. E. Comparison of Photon Stimulated Dissociation of Gas Phase and Chemisorbed CO. J. Chem. Phys. 1983, 78, 2998–3013. (32) Lambert, R.; Comrie, C. The Role of Primary and Secondary Electrons in Electron Induced Desorption and Dissociation: CO on Pt(111). Surf. Sci. 1973, 38, 197–209. (33) Lahtinen, J.; Vaari, J.; Kauraala, K. Adsorption and Structure Dependent Desorption of CO on Co(0001). Surf. Sci. 1998, 418, 502 – 510. (34) Gong, X.-Q.; Raval, R.; Hu, P. CO Dissociation and O Removal on Co(0001): a Density Functional Theory Study. Surf. Sci. 2004, 562, 247 – 256. (35) Allenspach, R.; Bischof, A. Magnetization Direction Switching in Fe/Cu(100) Epitaxial Films: Temperature and Thickness Dependence. Phys. Rev. Lett. 1992, 69, 3385–3388. 26


Page 26 of 31

Page 27 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(36) Kashuba, A.; Pokrovsky, V. L. Stripe Domain Structures in a Thin Ferromagnetic Film. Phys. Rev. Lett. 1993, 70, 3155–3158. (37) As can be seen in Figure 2f, the dichroic contrast recorded outside the irradiated disk is up to a factor ∼ 3.5 times larger than inside. This value matches the expected contrast difference for in-plane and out-of-plane magnetization for an incidence angle of the photon beam on the sample of 16◦ , as occurs in our set-up. Note also that the XMCD contrast in the stripe pattern is independent from the stripe orientation, which further confirms that the magnetization is oriented out-of-plane in the irradiated areas. (38) Wang, H.; Wong, S. P.; Cheung, W. Y.; Ke, N.; Wen, G. H.; Zhang, X. X.; Kwok, R. W. M. Magnetic Properties and Structure Evolution of Amorphous Co-C Nanocomposite Films Prepared by Pulsed Filtered Vacuum Arc Deposition. J. Appl. Phys. 2000, 88, 4919–4921. (39) Preobrajenski, A. B.; Ng, M. L.; Vinogradov, A. S.; M˚ artensson, N. Controlling Graphene Corrugation on Lattice-Mismatched Substrates. Phys. Rev. B 2008, 78, 073401. (40) G¨ unther, S.; Kolmakov, A.; Kovac, J.; Kiskinova, M. Artefact Formation in Scanning Photoelectron Emission Microscopy. Ultramicroscopy 1998, 75, 35 – 51. (41) Denley, D.; Perfetti, P.; Williams, R. S.; Shirley, D. A.; Stöhr, J. Carbon K-edge Fine Structure in Graphite Foils and in Thin-film Contaminants on Metal Surfaces. Phys. Rev. B 1980, 21, 2267–2273. (42) Chauvet, C.; Polack, F.; Silly, M. G.; Lagarde, B.; Thomasset, M.; Kubsky, S.; Duval, J. P.; Risterucci, P.; Pilette, B.; Yao, I.; Bergeard, N.; Sirotti, F. Carbon Contamination of Soft X-ray Beamlines: Dramatic Anti-reflection Coating Effects Observed in the 1keV Photon Energy Region. J. Synchrotron Radiat. 2011, 18, 761–764.

27



(43) Lahtinen, J.; Vaari, J.; Taio, A.; Vehanen, A.; Hautojrvi, P. Adsorption and Desorption Measurements of CO and O2 on Cobalt. Vacuum 1990, 41, 112–114. (44) Nowitzki, T.; Carlsson, A. F.; Martyanov, O.; Naschitzki, M.; Zielasek, V.; Risse, T.; Schmal, M.; Freund, H.-J.; Bäumer, M. Oxidation of Alumina-Supported Co and CoPd Model Catalysts for the Fischer-Tropsch Reaction. J. Phys. Chem. C 2007, 111, 8566–8572. (45) Gabaly, F. E.; McCarty, K. F.; Schmid, A. K.; de la Figuera, J.; Mu˜ noz, M. C.; Szunyogh, L.; Weinberger, P.; Gallego, S. Noble Metal Capping Effects on the Spinreorientation Transitions of Co/Ru(0001). New J. Phys. 2008, 10, 073024. (46) Rougemaille, N.; N’Diaye, A. T.; Coraux, J.; Vo-Van, C.; Fruchart, O.; Schmid, A. K. Perpendicular magnetic anisotropy of cobalt films intercalated under graphene. Appl. Phys. Lett. 2012, 101, 142403. (47) Usachov, D.; Fedorov, A.; Otrokov, M. M.; Chikina, A.; Vilkov, O.; Petukhov, A.; Rybkin, A. G.; Koroteev, Y. M.; Chulkov, E. V.; Adamchuk, V. K.; Grneis, A.; Laubschat, C.; Vyalikh, D. V. Observation of Single-Spin Dirac Fermions at the Graphene/Ferromagnet Interface. Nano Letters 2015, 15, 2396–2401. (48) Varykhalov, A.; Marchenko, D.; Sánchez-Barriga, J.; Scholz, M. R.; Verberck, B.; Trauzettel, B.; Wehling, T. O.; Carbone, C.; Rader, O. Intact Dirac Cones at Broken Sublattice Symmetry: Photoemission Study of Graphene on Ni and Co. Phys. Rev. X 2012, 2, 041017. (49) Kashuba, A. B.; Pokrovsky, V. L. Stripe domain structures in a thin ferromagnetic film. Phys. Rev. B 1993, 48, 10335–10344. (50) Chen, G.; Schmid, A. K. Imaging and Tailoring the Chirality of Domain Walls in Magnetic Films. Adv. Mater. 2015, 27, 5738–5743.

28


Page 28 of 31

Page 29 of 31 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(51) Chen, G.; Mascaraque, A.; N’Diaye, A. T.; Schmid, A. K. Room Temperature Skyrmion Ground State Stabilized through Interlayer Exchange Coupling. Appl. Phys. Lett. 2015, 106, 242404. (52) Boulle, O. et al. Room-temperature Chiral Magnetic Skyrmions in Ultrathin Magnetic Nanostructures. Nature Nanotech. 2016, 11, 449–454. (53) Chen, C. T.; Idzerda, Y. U.; Lin, H.-J.; Smith, N. V.; Meigs, G.; Chaban, E.; Ho, G. H.; Pellegrin, E.; Sette, F. Experimental Confirmation of the X-Ray Magnetic Circular Dichroism Sum Rules for Iron and Cobalt. Phys. Rev. Lett. 1995, 75, 152–155. (54) Coraux, J.; NDiaye, A. T.; Rougemaille, N.; Vo-Van, C.; Kimouche, A.; Yang, H.X.; Chshiev, M.; Bendiab, N.; Fruchart, O.; Schmid, A. K. Air-Protected Epitaxial Graphene/Ferromagnet Hybrids Prepared by Chemical Vapor Deposition and Intercalation. J. Phys. Chem. Lett. 2012, 3, 2059–2063. (55) Bairagi, K.; Bellec, A.; Repain, V.; Chacon, C.; Girard, Y.; Garreau, Y.; Lagoute, J.; Rousset, S.; Breitwieser, R.; Hu, Y.-C.; Chao, Y. C.; Pai, W. W.; Li, D.; Smogunov, A.; Barreteau, C. Tuning the Magnetic Anisotropy at a Molecule-Metal Interface. Phys. Rev. Lett. 2015, 114, 247203. (56) Vo-Van, C.; Kassir-Bodon, Z.; Yang, H.; Coraux, J.; Vogel, J.; Pizzini, S.; BayleGuillemaud, P.; Chshiev, M.; Ranno, L.; Guisset, V.; David, P.; Salvador, V.; Fruchart, O. Ultrathin Epitaxial Cobalt Films on Graphene for Spintronic Investigations and Applications. New Journal of Physics 2010, 12, 103040. (57) Manfrinato, V. R.; Stein, A.; Zhang, L.; Nam, C.-Y.; Yager, K. G.; Stach, E. A.; Black, C. T. Aberration-corrected Electron Beam Lithography at the One Nanometer Length Scale. Nano Lett. 2017, 17, 4562–4567. (58) Parschau, M.; Christmann, K. Cobalt on Rhenium(0001) an Example of Thermally Activated Layer Intermixing and Surface Alloying. Surf. Sci. 1999, 423, 303 – 323. 29



(59) de la Figuera, J.; Puerta, J.; Cerda, J.; Gabaly, F. E.; McCarty, K. Determining the Structure of Ru(0001) from Low Energy Electron Diffraction of a Single Terrace. Surf. Sci. 2006, 600, L105 – L109. (60) Gabaly, F. E.; Puerta, J. M.; Klein, C.; Saa, A.; Schmid, A. K.; McCarty, K. F.; Cerda, J. I.; de la Figuera, J. Structure and Morphology of Ultrathin Co/Ru(0001) Films. New J. Phys. 2007, 9, 80.

30


Page 30 of 31

Page 31 of 31 ACS Applied Materials & Interfaces 1 2 3 4 5 6


Magnetic Patterning by Electron Beam-Assisted Carbon Lithography

Recommend Documents